{"id":"W6891628535","doi":"10.48448/n5hs-z234","title":"CODER: An efficient framework for improving retrieval through COntextual Document Embedding Reranking","year":2022,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Thomson Reuters (Canada)","funders":"","keywords":"Embedding; Relevance (law); Document retrieval; Set (abstract data type); Ranking (information retrieval); Context (archaeology); Encoder; Training set","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.004957553,0.0009419288,0.0009609789,0.001267952,0.00205265,0.0009331701,0.002887254,0.0004885579,0.006007099],"category_scores_gemma":[0.002102839,0.0009422042,0.0002447537,0.002663376,0.001821148,0.0005910022,0.001198117,0.001390169,0.0003126487],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002604144,"about_ca_system_score_gemma":0.001770349,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004813517,"about_ca_topic_score_gemma":0.0002064851,"domain_scores_codex":[0.99065,0.0002323208,0.0009820396,0.002729083,0.003366737,0.002039828],"domain_scores_gemma":[0.9953848,0.0005824094,0.001359418,0.001910407,0.0003511031,0.0004119163],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001739122,0.002609742,0.0001665205,0.001282533,0.0005759728,0.0002093493,0.02340877,0.07663316,0.02555491,0.7229002,0.1064137,0.03850606],"study_design_scores_gemma":[0.004219104,0.002180648,0.000008976178,0.001102195,0.0004863491,0.0001053016,0.01277131,0.4643444,0.001901919,0.02251939,0.4860342,0.004326157],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004971419,0.00351455,0.8009166,0.0005558674,0.01158676,0.0109301,0.002977324,0.005639452,0.1589079],"genre_scores_gemma":[0.2025378,0.00003907099,0.7258282,0.001897129,0.004096264,0.0004306077,0.001088445,0.004942103,0.05914033],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7003808,"threshold_uncertainty_score":0.9993029,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03348975065760965,"score_gpt":0.3583281650714972,"score_spread":0.3248384144138875,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}